In PID adjustment support tools for generating models to be controlled and adjusting control parameters, target quantity data that are inputted in order to generate a model (time-series data of control-response waveforms and input data such as coordinate locations of way points, which serve as targets) and control-response waveforms that are calculated based on the models that are generated (hereinafter termed the model-response waveforms) are overlaid and graphed as modeling result to display the degree of matching, and the user is prompted to evaluate the suitability of the models produced. At this time, indicators corresponding to the differences between the target quantity data and the model-response waveforms are displayed as numeric values, such as evaluation function values of the modeling. (See Japanese Unexamined Patent Application Publication 2004-038428.) Additionally, common graphic display functions, such as enlarging or reducing the scales of the graphs, are also used in order to check the quality of matching. In any event, the user checks the misalignment of the target quantity data and the model-response waveform in the graph, or checks whether or not the indicator value is small, in order to evaluate the suitability of the model.
However, even the when the target quantity data itself for the modeling includes uncertainty, as disclosed in Japanese Unexamined Patent Application Publication 2006-127079, a perfect match between the target quantity data and the model-response waveform is not always the most suitable, conventionally attempts have been made to cause the target quantity data and the model-response waveform to match as closely as possible.
As described above, when the target quantity data itself for the modeling includes uncertainty, a perfect match between the target quantity data and the model-response waveform is not necessarily the most appropriate. Nevertheless, in the method according to the prior art wherein the model-response waveform is checked, if there is a misalignment between the target quantity data and the model-response waveform, then the user will attempt to match the target quantity data and the model-response waveform, and thus there is a problem in that the user falls into unnecessary trial-and-error by performing repetitive calculations and revisions to the model the equation, or is unable to evaluate the appropriateness of the model and thus is unable to move to the control parameter adjusting operation.
The present invention was created in order to solve the problems set forth above, and the object thereof is to provide a PID adjustment support device and method that reduces the likelihood of the pursuit of excessive accuracy by making it easy for the user to recognize the tolerance interval in the target quantity data that must be considered in advance when checking the target quantity data and the model-response waveform.